Dennis Glennon

Measuring the Default Risk of Small Business Loans: A Survival Analysis Approach

In this paper we report that, although medium-maturity loans originated under the SBA 7(a) loan guarantee program are targeted to small firms that fail to obtain credit through conventional channels, the default experience is comparable to that of a large percentage of loans held by larger commercial banks. We establish that the default behavior of these loans is time sensitive— as a loan seasons, the likelihood of default increases initially, peaks in the second year, and declines thereafter. Using a discrete-time hazard framework, we show that the likelihood of default is conditional on borrower, lender, and loan characteristics, and changes in economic conditions.

Evaluating Statistical Models of Mortgage Lending Discrimination: A Bank-Specific Analysis

We present our efforts to develop bank-specific models to test for the presence of mortgage lending discrimination. We discuss the potential for selection and simultaneity biases and delineate the conditions under which a single-equation model is appropriate. The results from three national banks demonstrate that, by incorporating the specific underwriting guidelines of each bank, our alternative approach significantly improves the ability of the model to explain the outcomes of the mortgage lending decision process when compared to a single generic specification applied across all banks. Our results also demonstrate the difficulties encountered in attempting to incorporate the specifics of a banks underwriting criteria and the remaining potential for omitted-variables problems.

The Information Revolution and Small Business Lending: The Missing Evidence

This paper provides empirical confirmation for Petersen and Rajans (2002) widely accepted conjecture that information technology was the primary driver of the observed increase in small business borrower-lender distances in the United States in recent years. Using a different data source for small business loans, we show that annual increases in borrower-lender distances were slow and steady prior to 1993 (the end point in Petersen and Rajans data) but accelerated rapidly after that. Importantly, we are able to assign at least half of this acceleration to the adoption of credit scoring technologies by the lending banks. Our tests also reveal strong statistical associations between lending distances and borrower characteristics, lender characteristics, market conditions, regulatory constraints, moral hazard incentives, and principal-agent incentives.

Warranty, quality and price in the US automobile market

The observed negative relationship between quality and warranty in the US auto market is analysed using a cost-based approach developed by Cooper and Ross. The theory of warranty and quality choice by producers and consumers is extended by endogenizing the joint price, quality and warranty decision. Differences in producer costs and consumer preferences are found to explain the inverse relationship, which suggests that American manufacturers should adjust their pricing, warranty and quality strategy when entering foreign markets.

Regional econometric models that reflect labor market relations

Abstract Regional economic forecasting is often hampered by a lack of reliable data. This study attempts to improve the efficiency of forecasts by incorporating interindustry linkages in wage and employment determination into an econometric model. Prior information of historical linkages is utilized in the form of restricted least squares and applied to the Louisville metropolitan area. We provide a step toward the specification of reliable models that reflect the local institutional framework which is often ignored in neoclassical, labor market-based model specifications. Comparisons on forecast accuracy with other structural models found in the literature favor the use of local institutional factors, though univariate time series models still tend to outperform the econometric approach.

Cost structures of banks grouped by strategic conduct

The hypothesis that the banking system consists of firms that use the same production technology is tested and rejected in this study. Six groupings of the population of commercial banks are identified using cluster analysis. The banks are grouped to reflect similar production technologies within groups but different technologies across groups as defined by the strategic conduct (i.e., activities) of the banks. The results suggest that banks in different clusters employ production processes that feature different degrees of substitutability between factors of production, and that the estimates of input substitutability for those groups look quite different from those estimated based on the full population of commercial banks. The impact of the homogeneity production technology assumption on the measurement of cost efficiency is also assessed. The results show that partitioning the industry by strategic conduct reduces the average inefficiency in the industry. These results support those found by others who used similar partitioning criteria but a more narrowly defined sample of banks.

Estimating the income, price, and interest elasticities of housing demand

Abstract There are many references in the housing market literature that suggest that the housing market contains characteristics of both stock and flows. However, most studies focus on either the stock or the flow aspect and ignore the simultaneous nature of the stock/flow concept. In this paper, the dual aspect of the housing market is formally incorporated using a stock/flow equilibrium model. The income, price, and interest elasticities are then estimated under the condition that both the flow excess demand and the market excess demand are zero.

Development and Validation of Credit Scoring Models

Accurate credit-granting decisions are crucial to the efficiency of the decentralized capital allocation mechanisms in modern market economies. Credit bureaus and many .nancial institutions have developed and used credit-scoring models to standardize and automate, to the extent possible, credit decisions. We build credit scoring models for bankcard markets using the Office of the Comptroller of the Currency, Risk Analysis Division (OCC/RAD) consumer credit database (CCDB). This unusu- ally rich data set allows us to evaluate a number of methods in common practice. We introduce, estimate, and validate our models, using both out-of-sample contempora- neous and future validation data sets. Model performance is compared using both separation and accuracy measures. A vendor-developed generic bureau-based score is also included in the model performance comparisons. Our results indicate that current industry practices, when carefully applied, can produce models that robustly rank-order potential borrowers both at the time of development and through the near future. However, these same methodologies are likely to fail when the the objective is to accurately estimate future rates of delinquency or probabilities of default for individual or groups of borrowers.

Financial innovation, new assets, and the behavior of money demand

Abstract Two fundamental changes in US banking regulations have affected the behavior of money demand (M1). The first authorized checkable deposit accounts paying explicit interest rates. The second allowed these rates to be market determined. The theoretical literature does not directly address the impact of these events, suggesting that they are primarily an empirical issue. However, the empirical literature has yet to agree on the impact of financial innovation on money demand; for example, several studies report an increase in the elasticity of money demand, several others report a decline. This paper uses a Lancaster-type choice model to analyze formally the expected impact of these two changes on the demand for money. The model derives specific conditions under which (i) the demand for money increases as new assets are introduced and (ii) the impact of either the introduction of new assets or the elimination of interest rate restrictions on the elasticity of money demand.

Work Profiles of Research Statisticians

Analysis of work profiles of 138 persons who had published in the Journal of the American Statistical Association showed this study group published more often than 128 members of the Association and tended to enjoy research, being less interested in the theoretical contribution of their work and more interested in the likelihood of work being published. Administrative and committee duties were impediments for those who published.